Behavioral data assists decisions: exploring the mental representation of digital-self

Zhang, Yixin; Cui, Lizhen; He, Wei; Lü, Xudong; Wang, Shipeng

doi:10.1108/ijcs-03-2021-0011

Cited by 5 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, to compare BERT with other models available in the HuggingFace Transformers library (RQ3), we experiment with two recent Transformer-based architectures: (1) DeBERTa [15] -a model [52], [40], [68], [38], [29], [7], [64], [55], [62], [18], [60], [53] [69], [32], [65], [58], [47] [3], [9] ML-1M* [14] 18 13 (72%) 3 (17%) 2 (11%) [52], [40], [68], [29], [7], [23], [55], [18], [60], [51], [42], [46], [28] [12], [62], [47] [64], [9] Yelp [2] 10 6 (60%) 4 (40%) 0 (0%) [69], [1], [3], [58], [47], [42] [12], [32], [65], [33] Steam that improves BERT with a disentangled attention mechanism [15] where each word is encoded using two vectors (a vector for content and a vector for position); (2) ALBERT [27] -a model that improves BERT via separating the size of the hidden state of the vocabulary embedding from t...…”

Section: Modelsmentioning

confidence: 99%

A Systematic Review and Replicability Study of BERT4Rec for Sequential Recommendation

Petrov

Macdonald

2022

Proceedings of the 16th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

BERT4Rec is an effective model for sequential recommendation based on the Transformer architecture. In the original publication, BERT4Rec claimed superiority over other available sequential recommendation approaches (e.g. SASRec), and it is now frequently being used as a state-of-the art baseline for sequential recommendation. However, not all subsequent publications confirmed its superiority and have proposed other models that were shown to outperform BERT4Rec in effectiveness. In this paper we systematically review all publications that compare BERT4Rec with another popular Transformer-based model, namely SASRec, and show that BERT4Rec results are not consistent within these publications. To understand the reasons behind this inconsistency, we analyse the available implementations of BERT4Rec and show that we fail to reproduce results of the original BERT4Rec publication when using their default configuration parameters. However, we are able to replicate the reported results with the original code if training for a much longer amount of time (up to 30x) compared to the default configuration. We also propose our own implementation of BERT4Rec based on the HuggingFace Transformers library, which we demonstrate replicates the originally reported results on 3 out 4 datasets, while requiring up to 95% less training time to converge. Overall, from our systematic review and detailed experiments, we conclude that BERT4Rec does indeed exhibit state-of-the-art effectiveness for sequential recommendation, but only when trained for a sufficient amount of time. Additionally, we show that our implementation can further benefit from adapting other Transformer architectures that are available in the HuggingFace Transformers library (e.g. using disentangled attention, as provided by DeBERTa, or larger hidden layer size cf. ALBERT). For example, on the MovieLens-1M dataset, we demonstrate that both these models can improve BERT4Rec performance by up to 9%. Moreover, we show that an ALBERT-based BERT4Rec model achieves better performance on that dataset than state-of-the-art results reported in the most recent publications.

show abstract

Section: Modelsmentioning

confidence: 99%

A Systematic Review and Replicability Study of BERT4Rec for Sequential Recommendation

Petrov

Macdonald

2022

Proceedings of the 16th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

show abstract

“…It has long been a fascinating topic that the structure of interaction networks has a significant influence on people's behavior. Behaviors [66,67] , minds [68] , and tendencies, such as smoking [69][70][71] , drinking [70] , role conflict [72] , and obesity [71] , have all been found to potentially influence individuals through their personal social networks. Therefore, many studies have proposed that in order to implement effective interventions on individual behaviors, the influence of neighbors in social networks needs to be fully considered [73,74] .…”

Section: Effect Of Network Structure On Human Behaviormentioning

confidence: 99%

Diminishing the Perception Bias in the Working Environment Using a Network Generation-Based Framework

Qian,

Zhang,

Sun

et al. 2024

Tsinghua Sci. Technol.

View full text Add to dashboard Cite

Income inequality is widespread in human society and has important implications for human behavior.People's perception of the environment bridges income inequality and individual behavioral decisions. Existing research suggests that social income inequality is usually biased, either overestimated or underestimated.However, such phenomena have not been fully explored with quantitive prove, especially in the working environment based on performance data of actual production. In fact, the correct perception of people is the basis of a fair environment for their production decisions. Thus, the perception bias may weaken the adaptability and competitiveness of a company in the market. This paper first confirms the prevalence of individual perception bias of income inequality within the working environment based on actual production data.Our results show that people tend to underestimate income inequity around them, and this underestimation grows with the real unfairness of the working environment. Further, this paper proposes a network generationbased framework with a three-layer structure to correct perception bias using a cooperative network reengineering approach. Within the framework, a homophily-based generative network model is proposed as the key algorithm. Our simulation results show that our proposed framework effectively reduces individuals' perception bias of income inequality.

show abstract

“…Early researchers try to utilize traditional machine learning methods, i.e. support vector machine (SVM) [11]- [14], naive bayesian (NB) [15] [16], random forest (RF) [17], k-nearest neighbor (KNN) [18], logistic regression (LR) [19] [20] and decision tree (DT) [21] for MDD recognition. Among these methods, feature extraction is crucial for the accuracy improvement of MDD recognition.…”

Section: Introductionmentioning

confidence: 99%

Multi-Granularity Graph Convolution Network for Major Depressive Disorder Recognition

Sun,

Xu,

Zhao

et al. 2024

IEEE Trans. Neural Syst. Rehabil. Eng.

Self Cite

View full text Add to dashboard Cite

Major depressive disorder (MDD) is the most common psychological disease. To improve the recognition accuracy of MDD, more and more machine learning methods have been proposed to mining EEG features, i.e. typical brain functional patterns and recognition methods that are closely related to depression using resting EEG signals. Most existing methods typically utilize threshold methods to filter weak connections in the brain functional connectivity network (BFCN) and construct quantitative statistical features of brain function to measure the BFCN.However, these thresholds may excessively remove weak connections with functional relevance, which is not conducive to discovering potential hidden patterns in weak connections. In addition, statistical features cannot describe the topological structure features and information network propagation patterns of the brain's different functional regions. To solve these problems, we propose a novel MDD recognition method based on a multi-granularity graph convolution network (MGGCN). On the one hand, this method applies multiple sets of different thresholds to build a multi-granularity functional neural network, which can remove noise while fully retaining valuable weak connections. On the other hand, this method utilizes graph neural network to learn the topological structure features and brain saliency patterns of changes between brain functional regions on the multi-granularity functional neural network. Experimental results on the benchmark datasets validate the superior performance and time complexity of MGGCN. The analysis shows that as the granularity increases, the connectivity defects in the right frontal(RF) and right temporal (RT) regions, left temporal(LT) and left posterior(LP) regions increase. The brain functional connections in these regions can serve as potential biomarkers for MDD recognition.

show abstract

Behavioral data assists decisions: exploring the mental representation of digital-self

Cited by 5 publications

References 26 publications

A Systematic Review and Replicability Study of BERT4Rec for Sequential Recommendation

A Systematic Review and Replicability Study of BERT4Rec for Sequential Recommendation

Diminishing the Perception Bias in the Working Environment Using a Network Generation-Based Framework

Multi-Granularity Graph Convolution Network for Major Depressive Disorder Recognition

Contact Info

Product

Resources

About